Mannose, glucosamine and inositol monophosphate inhibit the effects of insulin on lipogenesis. Further evidence for a role for inositol phosphate-oligosaccharides in insulin action.

The mechanism of insulin signalling is not yet understood in detail. Recently, a role for inositol phosphate (IP)-oligosaccharides as second messengers transmitting the insulin signal at the post-kinase level was proposed. To evaluate this hypothesis further, we studied whether IP-oligosaccharides isolated from 'haemodialysate' have insulin-like activity. We found that these compounds mimic, in a dose-dependent fashion, the following effects of insulin in adipocytes. (1) Lipogenesis. Incorporation of [3H]glucose into lipids (expressed in nmol/min per 10(6) cells): basal, 0.74 +/- 0.05; insulin (1 mu unit/ml), 4.43 +/- 0.21; IP-oligosaccharide (2 micrograms/ml), 4.07 +/- 0.19. (2) Inhibition of isoprenaline (isoproterenol) (1 microM)-stimulated cyclic AMP levels and lipolysis. Cyclic AMP (pmol/10(5) cells): basal 0.84 +/- 0.05; isoprenaline, 4.03 +/- 0.19; isoprenaline + insulin (200 mu units/ml), 2.06 +/- 0.7; isoprenaline + IP-oligosaccharides (2 micrograms/ml), 2.4 +/- 0.29. Inhibition of lipolysis (mumol of glycerol/mg of protein): isoprenaline (1 microM), 166 +/- 11; isoprenaline+insulin (150 mu units/ml), 53 +/- 3.5; isoprenaline+IP-oligosaccharides (2 micrograms/ml), 58 +/- 5. (3) Stimulation of 3-O-methylglucose transport; basal, 9 +/- 3%; insulin (1 mu unit/ml), 67 +/- 4%, IP-oligosaccharides (2 micrograms/ml), 54 +/- 2%. To identify the active molecules of the IP-oligosaccharide fraction, competition experiments were performed. IP-oligosaccharide effects on lipogenesis were blocked by inositol monophosphate, glucosamine and mannose. In contrast, these compounds did not inhibit IP-oligosaccharide effects on membrane-mediated functions (3-O-methylglucose transport, cyclic AMP levels, lipolysis). We also found that the effect of insulin on lipogenesis was blocked by mannose, glucosamine and inositol monophosphate, whereas the insulin effects on 3-O-methylglucose, cyclic AMP and lipolysis were unaffected. The following conclusions were reached. (1) IP-oligosaccharides mimic the major metabolic effects of insulin in adipocytes. This is consistent with the proposed role of IP-oligosaccharides as second messengers of certain insulin effects. (2) Mannose and glucosamine are functionally important sugar residues for the effect of IP-oligosaccharide on lipogenesis. (3) The observation that mannose, inositol monophosphate and glucosamine block the action of insulin of on lipogenesis supports a role of mannose- and glucosamine-containing IP-oligosaccharides as second messengers for this insulin effect.